Lab5_EXTRA- Interactive Graphs and Animations from the COVID-19 reporting data

Data for the lab

library(tidyverse)

## -- Attaching packages ----------------------------------------------- tidyverse 1.3.0 --

## v ggplot2 3.3.0     v purrr   0.3.3
## v tibble  2.1.3     v dplyr   0.8.5
## v tidyr   1.0.2     v stringr 1.4.0
## v readr   1.3.1     v forcats 0.5.0

## -- Conflicts -------------------------------------------------- tidyverse_conflicts() --
## x dplyr::filter() masks stats::filter()
## x dplyr::lag()    masks stats::lag()

library(lubridate)

## 
## Attaching package: 'lubridate'

## The following object is masked from 'package:base':
## 
##     date

time_series_confirmed_long <- read_csv(url("https://raw.githubusercontent.com/CSSEGISandData/COVID-19/master/csse_covid_19_data/csse_covid_19_time_series/time_series_covid19_confirmed_global.csv")) %>%
  rename(Province_State = "Province/State", Country_Region = "Country/Region")  %>% 
               pivot_longer(-c(Province_State, Country_Region, Lat, Long),
                             names_to = "Date", values_to = "Confirmed") 

## Parsed with column specification:
## cols(
##   .default = col_double(),
##   `Province/State` = col_character(),
##   `Country/Region` = col_character()
## )

## See spec(...) for full column specifications.

# Let's get the times series data for deaths
time_series_deaths_long <- read_csv(url("https://raw.githubusercontent.com/CSSEGISandData/COVID-19/master/csse_covid_19_data/csse_covid_19_time_series/time_series_covid19_deaths_global.csv")) %>%
  rename(Province_State = "Province/State", Country_Region = "Country/Region")  %>% 
  pivot_longer(-c(Province_State, Country_Region, Lat, Long),
               names_to = "Date", values_to = "Deaths")

## Parsed with column specification:
## cols(
##   .default = col_double(),
##   `Province/State` = col_character(),
##   `Country/Region` = col_character()
## )
## See spec(...) for full column specifications.

# Create Keys 
time_series_confirmed_long <- time_series_confirmed_long %>% 
  unite(Key, Province_State, Country_Region, Date, sep = ".", remove = FALSE)
time_series_deaths_long <- time_series_deaths_long %>% 
  unite(Key, Province_State, Country_Region, Date, sep = ".") %>% 
  select(Key, Deaths)
# Join tables
time_series_long_joined <- full_join(time_series_confirmed_long,
    time_series_deaths_long, by = c("Key")) %>% 
    select(-Key)
# Reformat the data
time_series_long_joined$Date <- mdy(time_series_long_joined$Date)
# Create Report table with counts
time_series_long_joined_counts <- time_series_long_joined %>% 
  pivot_longer(-c(Province_State, Country_Region, Lat, Long, Date),
               names_to = "Report_Type", values_to = "Counts")

Controlling graph size in R markdown

In the opening line of the RMarkdown code chunk {r} you can control the output of the code, graphs, tables using knitr syntax. For example if {r, eval = FALSE} the code will not be run, but will be shown. If {r, code = FALSE} the code will not be shown, but will be run and the output will be shown (useful in reports where the reader is only interested in the results/graphs, but not the code). You can also suppress error messages and warnings so that the reader isn’t bothered by them (but you should take notice).

The dimensions of an individual graph in the RMarkdown document be adjusted by specifying the graph dimensions {r, fig.width = 6, fig.height = 6}.

Graphic output

You can export plots using the Export button in the Plots window. You can save as a pdf, svg, tiff, png, bmp, jpeg and eps. You can also write the output directly to a file. This is particularly useful for controling the final dimensions in a reproducible way and for manuscripts.

# Plot graph to a pdf outputfile
pdf("images/time_series_example_plot.pdf", width=6, height=3)
time_series_long_joined %>% 
  group_by(Country_Region,Date) %>% 
  summarise_at(c("Confirmed", "Deaths"), sum) %>% 
  filter (Country_Region == "US") %>% 
    ggplot(aes(x = Date,  y = Deaths)) + 
    geom_point() +
    geom_line() +
    ggtitle("US COVID-19 Deaths")
dev.off()

## png 
##   2

# Plot graph to a png outputfile
ppi <- 300
png("images/time_series_example_plot.png", width=6*ppi, height=6*ppi, res=ppi)
time_series_long_joined %>% 
  group_by(Country_Region,Date) %>% 
  summarise_at(c("Confirmed", "Deaths"), sum) %>% 
  filter (Country_Region == "US") %>% 
    ggplot(aes(x = Date,  y = Deaths)) + 
    geom_point() +
    geom_line() +
    ggtitle("US COVID-19 Deaths")
dev.off()

## png 
##   2

RMarkdown loading images

Sometimes it is useful in controling the image layout for a report to file with the graph and then subsequently load it into the .Rmd file. This works with png files, but not pdfs. You can also upload images made with other bioinformatic tools into your RMarkdown report.

This is the RMarkdown style for inserting images. Your image must be in your working directory. This command is put OUTSIDE the r code chunk.

US COVID-19 Deaths

This is an alternative way using html. Remember that it must be in your working directory or you will need to specify the full path. The html is put OUTSIDE the r code chunk.

Another way to present a graph without the code is adding echo = FALSE within the r{} chunk - {r echo = FALSE}. This prevents code, but not the results from appearing in the knitr file.

Interactive graphs

With plotly/ggplotly (https://plot.ly/ggplot2/) you can make interactive graphs in your lab report.

There are two common formats used in graphing that you may come across in examples. One very common format that we have not used to date is putting your graph into a variable and then plotting the variable. They both work perfectly fine.

# Version 2
library(plotly)

## Warning: package 'plotly' was built under R version 3.6.3

## 
## Attaching package: 'plotly'

## The following object is masked from 'package:ggplot2':
## 
##     last_plot

## The following object is masked from 'package:stats':
## 
##     filter

## The following object is masked from 'package:graphics':
## 
##     layout

ggplotly(
  time_series_long_joined %>% 
    group_by(Country_Region,Date) %>% 
    summarise_at(c("Confirmed", "Deaths"), sum) %>% 
    filter (Country_Region == "US") %>% 
    ggplot(aes(x = Date,  y = Deaths)) + 
      geom_point() +
      geom_line() +
      ggtitle("US COVID-19 Deaths")
 )

library(plotly)
US_deaths <- time_series_long_joined %>% 
    group_by(Country_Region,Date) %>% 
    summarise_at(c("Confirmed", "Deaths"), sum) %>% 
    filter (Country_Region == "US")
 p <- ggplot(data = US_deaths, aes(x = Date,  y = Deaths)) + 
        geom_point() +
        geom_line() +
        ggtitle("US COVID-19 Deaths")
ggplotly(p)

Animated graphs with gganimate

Animated graphs when done right have a great visual impact. You can do this in R and have your animations embedded on your web page. Essentially gganimate creates a series of files that are encompassed in a gif file. In addition to having this gif as part of your report file, you can save the gif and use in a slide or other presentations. It just takes a few lines of code to covert and existing ggplot graph into an animation.

These are some important gganimate functions:

• __transition_*()__ defines how the data should be spread out and how it relates to itself across time.
• __view_*()__ defines how the positional scales should change along the animation.
• __shadow_*()__ defines how data from other points in time should be presented in the given point in time.
• enter_()/exit_() defines how new data should appear and how old data should disappear during the course of the animation.
• ease_aes() defines how different aesthetics should be eased during transitions.

Creating the animations

Below are the packages I installed. There may be others that you need, in particular to rendering gifs. Several people needed to install the packages gifski and av Some of the examples may take several minutes to create the animation.

library(gganimate)
library(transformr)
theme_set(theme_bw())

An animation of the confirmed cases in select countries.

data_time <- time_series_long_joined %>% 
    group_by(Country_Region,Date) %>% 
    summarise_at(c("Confirmed", "Deaths"), sum) %>% 
    filter (Country_Region %in% c("China","Korea, South","Japan","Italy","US")) 
p <- ggplot(data_time, aes(x = Date,  y = Confirmed, color = Country_Region)) + 
      geom_point() +
      geom_line() +
      ggtitle("Confirmed COVID-19 Cases") +
      geom_point(aes(group = seq_along(Date))) +
      transition_reveal(Date) 
# Some people needed to use this line instead
# animate(p,renderer = gifski_renderer(), end_pause = 15)
animate(p, end_pause = 15)

Exercises

Wednesday labs are supplemental. The only requirement for students in 497 is that you reproduce the above code on your computer and add it to your GitHub site. Students in 697 need to complete the following challanges.

Challenge 1

Print a graph (different from the one above) to a png file using 3*ppi for the height and width and display the png file in the report using the above R Markdown format.

Printing graph from Lab 5 Exercise 4 as a png file.

time_series_deaths <- read_csv("data/time_series_covid19_deaths_global.csv")%>%
  rename(Province_State = "Province/State", Country_Region = "Country/Region")

## Parsed with column specification:
## cols(
##   .default = col_double(),
##   `Province/State` = col_character(),
##   `Country/Region` = col_character()
## )

## See spec(...) for full column specifications.

time_series_deaths_long <- time_series_deaths %>% 
               pivot_longer(-c(Province_State, Country_Region, Lat, Long),
                            names_to = "Date", values_to = "Deaths")

time_series_deaths_long$Date = mdy(time_series_deaths_long$Date)

ppi <- 300
png("images/time_series_challenge1_plot.png", width=3*ppi, height=3*ppi, res=ppi)
time_series_deaths_long %>% 
  group_by(Date) %>% 
  summarize(Sum = sum(Deaths)) %>% 
    ggplot(aes(x = Date, y = Sum)) +
    geom_point() +
    ggtitle("COVID-19 Deaths Worldwide") +
    labs(x = "Date", y = "Total Confirmed Deaths")
dev.off()

## png 
##   2

US COVID-19 Worldwide Deaths

Challenge 2

Turn one of the exercises from Lab 5 into an interactive graph with plotyly.

Using graphs from Lab 5 Exercise 4

ggplotly(
  time_series_deaths_long %>% 
  group_by(Date) %>% 
  summarize(Sum = sum(Deaths)) %>% 
    ggplot(aes(x = Date, y = Sum)) +
    geom_point() +
    ggtitle("COVID-19 Deaths Worldwide") +
    labs(x = "Date", y = "Total Confirmed Deaths")
)

Challenge 3

Create an animated graph of your choosing using the time series data to display an aspect (e.g. states or countries) of the data that is important to you.

Using graph from Lab 5 Exercise 7-

time_series_deaths <- read_csv("data/time_series_covid19_deaths_global.csv")%>%
  rename(Province_State = "Province/State", Country_Region = "Country/Region")

## Parsed with column specification:
## cols(
##   .default = col_double(),
##   `Province/State` = col_character(),
##   `Country/Region` = col_character()
## )

## See spec(...) for full column specifications.

time_series_deaths_long <- time_series_deaths %>% 
               pivot_longer(-c(Province_State, Country_Region, Lat, Long),
                            names_to = "Date", values_to = "Deaths") 

Top_10 = time_series_deaths_long %>% 
  group_by(Country_Region) %>% 
  summarize(T_Deaths = sum(Deaths)) %>% 
  arrange(desc(T_Deaths)) %>% 
  slice(1:10)

Top_10_time_series_deaths_long = time_series_deaths_long[which(time_series_deaths_long$Country_Region %in% Top_10$Country_Region),]

Top_10_time_series_deaths_long$Date = mdy(Top_10_time_series_deaths_long$Date)

plot_dat = Top_10_time_series_deaths_long %>%
  group_by(Country_Region, Date) %>% 
  summarize(T_DeathsbyDate = sum(Deaths))

p = ggplot(plot_dat, aes(x = Date, y = T_DeathsbyDate, color = Country_Region)) +
    geom_point() +
    geom_line() +
    ggtitle("COVID-19 Deaths in Top 10 Countries Worldwide") +
    labs(x = "Date", y = "Deaths") +
    geom_point(aes(group = seq_along(Date))) +
    transition_reveal(Date)

animate(p, renderer = gifski_renderer(), end_pause = 15)